Back to Search Start Over

Debiasing Word Embeddings with Nonlinear Geometry

Authors :
Cheng, Lu
Kim, Nayoung
Liu, Huan
Publication Year :
2022

Abstract

Debiasing word embeddings has been largely limited to individual and independent social categories. However, real-world corpora typically present multiple social categories that possibly correlate or intersect with each other. For instance, "hair weaves" is stereotypically associated with African American females, but neither African American nor females alone. Therefore, this work studies biases associated with multiple social categories: joint biases induced by the union of different categories and intersectional biases that do not overlap with the biases of the constituent categories. We first empirically observe that individual biases intersect non-trivially (i.e., over a one-dimensional subspace). Drawing from the intersectional theory in social science and the linguistic theory, we then construct an intersectional subspace to debias for multiple social categories using the nonlinear geometry of individual biases. Empirical evaluations corroborate the efficacy of our approach. Data and implementation code can be downloaded at https://github.com/GitHubLuCheng/Implementation-of-JoSEC-COLING-22.<br />Comment: Accepted to COLING2022

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2208.13899
Document Type :
Working Paper